The long term objective of this program project is to develop a validated typology of the developmental language disorders (DLD) and autistic spectrum disorders (ASD) of preschool children. The focus during the first three years of the project was on identifying the most powerful measures to distinguish DLD and ASD from mental deficiency uncomplicated by autistic features (NAMD), and to identify subtypes of DLD and AMD, drawing from linguistic, neuropsychological, neurological, and historical perspectives. The aim of the current project is to follow longitudinally at ages 7 and 9 years some 340 DLD and ASD children subtyped during the first three years of this project in order: 1) to determine subtype stability or change over time: 2) to relate subtypes identified during the preschool years to cognitive status, language ability, academic achievement, social-emotional adjustment, and symptomatology at school-age. 3) to identify preschool factors prognostic of school-age language, academic, and social-emotional development and of diagnostic status of outcome. In addition, prototypical DLD and ASD children from the longitudinal cohort will serve as subjects in the morphometric neuroimaging, electrophysiologic and genetic external validating studies to be pursued as part of the program project. Our methodologic approach is clinical-empirical. Clinical diagnoses based on the impressions of the investigators evaluating the children live, performing standardized medical, neurologic, and psychiatric examinations, the responses of parents and teachers to questionnaires, and analysis of spontaneous language samples, sociability, and play are entered into a common data base. The data base also comprises results of standardized tests of language and cognition as well as historical and demographic data. Solutions based on clustering and other multivariate analyses of the data base undergo internal validating procedures to ensure their reliability are compared with the clinical impressions to assess the convergence of the clinical and empirical typologies.